Oulook: More statistical fog

POLITICIANS WILL only order a change in the way economic statistics are compiled if they think it will reflect them in a better light. For years, Gordon Brown, the Chancellor, has been complaining that the national statistics fail adequately to measure the growth in the output of the public sector. The result is the Atkinson Review, whose final report together with its 54 recommendations was published yesterday.

Some of these recommendations have already been incorporated by the Office for National Statistics into the way the figures are compiled - the ones that pertain to the National Health Service. To the delight of ministers, the effect was nearly to double the output of the NHS over the three years to the end of 2003, allowing the Government to announce worthwhile increases in the overall rate of GDP growth for each of these years as well.

But the process doesn't always go the Government's way. Productivity growth in the NHS was actually lower under the new methodology than the old, prompting the Health Secretary, John Reid, to describe the figures as "absurd" because in his view they failed to take proper account of improvements in the quality of care. No wonder he was rattled. The new statistics seemed to confirm what the sceptics were saying; the more money the Government spends, the less efficient the public sector becomes. That's not at all what the doctor had ordered.

The ONS announced yesterday that it would be incorporating a further nine of the report's final recommendations into the figures, while continuing to assess and consult on the others. These all relate to measuring educational output, where there are similar concerns to those expressed over the NHS. For instance, if government spending is causing class sizes to get smaller, does that count as an increase or decrease in output and productivity?

The revisions will be ready for publication by the end of April but that doesn't necessarily mean they will be published then. The imminence of the election makes it all very political. If the result is good news for the Government, then Len Cook, head of the ONS, would be accused of deliberately trying to boost the Government's electoral prospects by publishing. Likewise he could be accused of burying the bad news if he fails to publish statistics that show extra government spending on education in a poor light. Now who was it who said there are lies, damned lies and statistics?